Ad Hoc Networks for Disaster Scenarios and/or Threat Detectiondownloads.hindawi.com/journals/specialissues/967324.pdf · Ad Hoc Networks for Disaster Scenarios and/or Threat Detection

International Journal of Distributed Sensor Networks

Ad Hoc Networks for Disaster Scenarios and/or Threat Detection

Guest Editors: Federico Barrero, Sergio Toral, Princy Johnson, Mesut Gnes, and Eleana Asimakopoulou

Ad Hoc Networks for Disaster Scenarios

and/or Threat Detection

International Journal of Distributed Sensor Networks

Ad Hoc Networks for Disaster Scenarios

and/or Threat Detection

Guest Editors: Federico Barrero, Sergio Toral,

Princy Johnson, Mesut Gunes, and Eleana Asimakopoulou

Copyright 2015 Hindawi Publishing Corporation. All rights reserved.

is is a special issue published in International Journal of Distributed Sensor Networks. All articles are open access articles distributedunder the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, pro-vided the original work is properly cited.

Editorial Board

Jemal H. Abawajy, AustraliaMiguel Acevedo, USACristina Alcaraz, SpainAna Alejos, SpainMohammod Ali, USAGiuseppe Amato, ItalyHabib M. Ammari, USAMichele Amoretti, ItalyChristos Anagnostopoulos, UKLi-Minn Ang, AustraliaNabil Aouf, UKFrancesco Archetti, ItalyMasoud Ardakani, CanadaMiguel Ardid, SpainMuhammad Asim, UKStefano Avallone, ItalyJose L. Ayala, SpainJavier Bajo, SpainN. Balakrishnan, IndiaPrabir Barooah, USAFederico Barrero, SpainPaolo Barsocchi, ItalyPaolo Bellavista, ItalyOlivier Berder, FranceRoc Berenguer, SpainJuan A. Besada, SpainGennaro Boggia, ItalyAlessandro Bogliolo, ItalyEleonora Borgia, ItalyJanos Botzheim, JapanFarid Boussaid, AustraliaArnold K. Bregt, NetherlandsRichard R. Brooks, USATed Brown, USADavide Brunelli, ItalyJames Brusey, UKCarlos T. Calafate, SpainTiziana Calamoneri, ItalyJose Camacho, SpainJuan Carlos Cano, SpainXianghui Cao, USAJoao Paulo Carmo, BrazilRoberto Casas, SpainLuca Catarinucci, ItalyMichelangelo Ceci, Italy

Yao-Jen Chang, TaiwanNaveen Chilamkurti, AustraliaWook Choi, KoreaHyunseung Choo, KoreaKim-Kwang R. Choo, AustraliaChengfu Chou, TaiwanMashrur A. Chowdhury, USATae-Sun Chung, KoreaMarcello Cinque, ItalySesh Commuri, USAMauro Conti, ItalyAlfredo Cuzzocrea, ItalyDonatella Darsena, ItalyDinesh Datla, USAAmitava Datta, AustraliaIyad Dayoub, FranceDanilo De Donno, ItalyLuca De Nardis, ItalyFloriano De Rango, ItalyPaula de Toledo, SpainMarco Di Felice, ItalySalvatore Distefano, ItalyLongjun Dong, ChinaNicola Dragoni, DenmarkGeorge P. Ehymoglou, GreeceFrank Ehlers, ItalyMelike Erol-Kantarci, CanadaFarid Farahmand, USAMichael Farmer, USAFlorentino Fdez-Riverola, SpainGianluigi Ferrari, ItalySilvia Ferrari, USAGiancarlo Fortino, ItalyLuca Foschini, ItalyJean Y. Fourniols, FranceDavid Galindo, SpainEnnio Gambi, ItalyWeihua Gao, USAA.-Javier Garca-Sanchez, SpainPreetam Ghosh, USAAthanasios Gkelias, UKIqbal Gondal, AustraliaFrancesco Grimaccia, ItalyJayavardhana Gubbi, AustraliaSong Guo, Japan

Andrei Gurtov, FinlandMohamed A. Haleem, USAKijun Han, Republic of KoreaQi Han, USAZdenek Hanzalek, Czech RepublicShinsuke Hara, JapanWenbo He, CanadaPaul Honeine, FranceFeng Hong, ChinaChin-Tser Huang, USAHaiping Huang, ChinaXinming Huang, USAMohamed Ibnkahla, CanadaSyed K. Islam, USALillykutty Jacob, IndiaWon-Suk Jang, KoreaAntonio Jara, SwitzerlandShengming Jiang, ChinaYingtao Jiang, USANing Jin, ChinaRaja Jurdak, AustraliaKonstantinos Kalpakis, USAIbrahim Kamel, UAEJoarder Kamruzzaman, AustraliaRajgopal Kannan, USAJohannes M. Karlsson, SwedenGour C. Karmakar, AustraliaMarcos D. Katz, FinlandJamil Y. Khan, AustraliaSherif Khattab, EgyptHyungshin Kim, Republic of KoreaSungsuk Kim, Republic of KoreaAndreas Konig, GermanyGurhan Kucuk, TurkeySandeep S. Kumar, NetherlandsJuan A. L. Riquelme, SpainYee W. Law, AustraliaAntonio Lazaro, SpainDidier Le Ruyet, FranceJoo-Ho Lee, JapanSeokcheon Lee, USAYong Lee, USAStefano Lenzi, ItalyPierre Leone, SwitzerlandShancang Li, UK

Shuai Li, USAQilian Liang, USAWeifa Liang, AustraliaYao Liang, USAI-En Liao, TaiwanJiun-Jian Liaw, TaiwanAlvin S. Lim, USAAntonio Liotta, NetherlandsDonggang Liu, USAHai Liu, Hong KongYonghe Liu, USALeonardo Lizzi, FranceJaime Lloret, SpainKenneth J. Loh, USAJuan Carlos Lopez, SpainManel Lopez, SpainPascal Lorenz, FranceChun-Shien Lu, TaiwanJun Luo, SingaporeMichele Magno, ItalySabato Manfredi, ItalyAthanassios Manikas, UKPietro Manzoni, Spain

Alvaro Marco, SpainJose R. Martinez-de Dios, SpainAhmed Mehaoua, FranceNirvana Meratnia, NetherlandsChristian Micheloni, ItalyLyudmila Mihaylova, UKPaul Mitchell, UKMihael Mohorcic, SloveniaJose Molina, SpainAntonella Molinaro, ItalyJose I. Moreno, SpainSalvatore Morgera, USAKazuo Mori, JapanLeonardo Mostarda, ItalyV. Muthukkumarasamy, AustraliaKamesh Namuduri, USAAmiya Nayak, CanadaGeorge Nikolakopoulos, Sweden

Alessandro Nordio, ItalyMichael J. OGrady, IrelandGregory OHare, IrelandGiacomo Oliveri, ItalySaeed Olyaee, IranLuis Orozco-Barbosa, SpainSuat Ozdemir, TurkeyVincenzo Paciello, ItalySangheon Pack, Republic of KoreaMarimuthu Palaniswami, AustraliaMeng-Shiuan Pan, TaiwanSeung-Jong Park, USAMiguel A. Patricio, SpainLuigi Patrono, ItalyRosa A. Perez-Herrera, SpainPedro Peris-Lopez, SpainJanez Pers, SloveniaDirk Pesch, IrelandShashi Phoha, USARobert Plana, FranceCarlos Pomalaza-Raez, FinlandNeeli R. Prasad, DenmarkAntonio Puliato, ItalyHairong Qi, USAMeikang Qiu, USAVeselin Rakocevic, UKNageswara S.V. Rao, USALuca Reggiani, ItalyEric Renault, FranceJoel Rodrigues, PortugalPedro P. Rodrigues, PortugalLuis Ruiz-Garcia, SpainMohamed Saad, UAEStefano Savazzi, ItalyMarco Scarpa, ItalyArunabha Sen, USAOlivier Sentieys, FranceSalvatore Serrano, ItalyZhong Shen, ChinaChin-Shiuh Shieh, TaiwanMinho Shin, Korea

Pietro Siciliano, ItalyOlli Silven, FinlandHichem Snoussi, FranceGuangming Song, ChinaAntonino Staiano, ItalyMuhammad A. Tahir, PakistanJindong Tan, USAShaojie Tang, USALuciano Tarricone, ItalyKerry Taylor, AustraliaSameer S. Tilak, USAChuan-Kang Ting, TaiwanSergio Toral, SpainVicente Traver, SpainIoan Tudosa, ItalyAnthony Tzes, GreeceBernard Uguen, FranceFrancisco Vasques, PortugalKhan A. Wahid, CanadaAgustinus B. Waluyo, AustraliaHonggang Wang, USAJianxin Wang, ChinaJu Wang, USAYuWang, USAomas Wettergren, USARan Wol, IsraelChase Wu, USANa Xia, ChinaQin Xin, Faroe IslandsChun J. Xue, Hong KongYuan Xue, USAGeng Yang, Chinaeodore Zahariadis, GreeceMiguel A. Zamora, SpainHongke Zhang, ChinaXing Zhang, ChinaJiliang Zhou, ChinaTing L. Zhu, USAXiaojun Zhu, ChinaYifeng Zhu, USADaniele Zonta, Italy

Contents

Ad Hoc Networks for Disaster Scenarios and/orreat Detection, Federico Barrero, Sergio Toral,Princy Johnson, Mesut Gunes, and Eleana AsimakopoulouVolume 2015, Article ID 949214, 2 pages

A Decision-Aided Situation Awareness Mechanism Based onMultiscale Dynamic Trust, Fangwei Li,Yifang Nie, Jiang Zhu, Haibo Zhang, and Fan LiuVolume 2015, Article ID 107921, 14 pages

A Survey on Multihop Ad Hoc Networks for Disaster Response Scenarios, D. G. Reina, M. Askalani,S. L. Toral, F. Barrero, E. Asimakopoulou, and N. BessisVolume 2015, Article ID 647037, 16 pages

A Novel Approach to Building a Heterogeneous Emergency Response Communication System,Janez Sterle, Miha Rugelj, Urban Sedlar, Luka Korsic, Andrej Kos, Peter Zidar, and Mojca VolkVolume 2015, Article ID 685253, 9 pages

Distributed Forest Fire Monitoring UsingWireless Sensor Networks, M. Angeles Serna, Rafael Casado,Aurelio Bermudez, Nuno Pereira, and Stefano TenninaVolume 2015, Article ID 964564, 18 pages

Implementation and Analysis of Clustering Techniques Applied on Pocket Switched Network,Muhammad Ali, Mah-Rukh Fida, Ameer Shakayb Arsalaan, and Awais AdnanVolume 2015, Article ID 239591, 6 pages

EditorialAd Hoc Networks for Disaster Scenarios and/or Threat Detection

Federico Barrero,1 Sergio Toral,1 Princy Johnson,2

Mesut Gnes,3 and Eleana Asimakopoulou4

1Department of Electronic Engineering, University of Seville, 41092 Sevilla, Spain2School of Engineering, Technology and Maritime Operations, Liverpool John Moores University, Liverpool L3 3AF, UK3Communication and Networked Systems (ComSys), Institute of Computer Science, University of Munster, 48149 Munster, Germany4University of Derby, Derby DE22 1GB, UK

Correspondence should be addressed to Federico Barrero; [email protected]

Received 13 September 2015; Accepted 13 September 2015

Copyright 2015 Federico Barrero et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Disaster scenarios and threat detection systems presentinteresting niche applications for ad hoc and wireless sensornetworks that can be used to showcase novel communicationtechniques to deal with unexpected conditions emergingfrom such scenarios.There is an opportunity to develop tech-niques suitable for the 21st century, where peoplemostly com-municate with each other using smart mobile devices such asiPhones. Communication among victims and crewmembersinvolved in rescue operations is crucial in order to alleviatedisaster consequences and save lives, for example, enablinga quick response from security forces to threat detectionsituations. However, cellular-based communicationsmay notbe possible after a disaster due to the damages caused tothe telecommunication infrastructure, leaving many peopleisolated and unprotected, and therefore having catastrophicconsequences for humanity.

The scope of the special issue is to encourage newadvances and developments in this field, while showcasingthe recent developments to the scientific community. Thepresented survey in the field, which addresses the applica-bility of such multihop ad hoc networks in disaster responsescenarios, will surely help new researchers to assimilate theexpertize level and rapidly contribute to the field with newdevelopments. Adetailed description of the existingmultihopad hoc network paradigms applicable to disaster responsescenarios is discussed within the work, where types of com-munications, wireless technologies, and research directionsin the field are analyzed. A thorough review of the existing

work for each ad hoc paradigm is also introduced in thesurvey paper, where the necessity of interoperability amongdifferent ad hoc paradigms is highlighted. Finally, a review ofthe future open challenges for multihop ad hoc networks indisaster response scenarios is strategically shown to start newdevelopments in the area.

The experiences and platforms case studies, demos, andprototype testing systems based on sensor networks includedin the special issue will show the utility of the wireless sensornetworks when managing disaster scenarios, which can beused in forest fire fighting operations to propose ways forapproximating the actual shape of the fire, assuming efficientfire detection dissemination layer in charge of broadcastingfire detection events. A similar proposal discusses a real-world pilot implementation customized for fire fighter ser-vices in Slovenia, where a survivable heterogeneous emer-gency communication system is proposed that combinesprofessional and commercial off-the-shelf equipment thatcommunicates over mobile and satellite links. The system isdeveloped to support public safety agencies in their day-to-day operation and disaster relief missions.

Ad hoc communication paradigms are also stated, andnew research ideas are introduced to the research communityin the real-time data collection and decisionmaking, securityand privacy, and system integrity and reliability fields, whensensor networks are used in disaster scenarios and/or threatdetection. Also, the existing clustering methods, which canbe considered an important tool for efficient communication

Hindawi Publishing CorporationInternational Journal of Distributed Sensor NetworksVolume 2015, Article ID 949214, 2 pageshttp://dx.doi.org/10.1155/2015/949214

http://dx.doi.org/10.1155/2015/949214

2 International Journal of Distributed Sensor Networks

in an intermittent Pocket Switched Networks, are evaluatedto minimize communication overhead and then modified toensure communication links between disjoint nodes.

An interesting solution for distributed computing trustis also presented as an effective method to minimize threatsin wireless sensor networks, where trust and satisfaction arebuilt based on Ebbinghaus forgetting regular and spatial cor-relations. Then, misbehaviors can be effectively suppressedin the network security area, and the proposed method canrecognize and handle entity attacks more effectively thanother mechanisms.

We hope that the special issue can stimulate furtherresearch interests in the ad hoc networks field, with particularapplication to the disaster scenarios and/or threat detectionareas, wheremuchmorework is for sure expected in the com-ing years.

Federico BarreroSergio Toral

Princy JohnsonMesut Gunes

Eleana Asimakopoulou

Research ArticleA Decision-Aided Situation Awareness Mechanism Based onMultiscale Dynamic Trust

Fangwei Li, Yifang Nie, Jiang Zhu, Haibo Zhang, and Fan Liu

Chongqing Key Lab of Mobile Communications Technology, Chongqing University of Posts and Telecommunications,Chongqing 400065, China

Correspondence should be addressed to Yifang Nie; [email protected]

Received 25 December 2014; Revised 8 March 2015; Accepted 15 March 2015

Academic Editor: Sergio Toral

Copyright 2015 Fangwei Li et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Attacks are always seekingways of exploiting any existingweakness inwireless network.The purpose of security situation awarenessis to recognize, analyze, forecast, and handle the misbehaviors, assisting network management. Nevertheless, the appearance ofubiquitous network and the network convergence technology has made a challenge to realize network security and adaptation.Aiming at these research problems, this paper proposes a decision-aided situation awareness mechanism based on multiscaledynamic trust from the perspective of time and space, which can recognize misbehaviors and regard social network as the researchobject. We build trust and satisfaction based on Ebbinghaus forgetting regular and spatial correlations.This mechanism carries outdecision making assessment through trust authenticity test, logicality test, and feedback parameters. In addition, load balance isused to avoid resource congestion. Simulation analysis demonstrates that compared with other trust mechanisms, this mechanismproposed in this paper can recognize and handle entity attacks more effectively, which is relatively eclectic and realistic in aspect oftrust mensuration.

1. Introduction

Security situation awareness [1], including security elementsdetection, perception, prediction, evaluation, visualization,and management, can sense security information, extractingand analyzing security essential elements. It can also find outthe existent threats and attacks [2], to cope with the char-acteristics of dynamics, openness, and uncertainty. Simul-taneously, that also contains the steps of decision support,risk assessment, and situation visualization. Security situationawareness is increasingly becoming popular with the appear-ance of network convergence technology. Those networkentities of situation awareness mechanism, which can flexiblyand intelligently respond to the environment change, can beregarded as interactive entities existing in multiagent system,Peer to Peer Network named P2P, Wireless Sensor Networknamed WSN, and Mobile Ad hoc Network named MANET.Nevertheless, attacks and existing weaknesses also havemadea challenge for researchers to effect network security andadaptation with the dynamic change of location, permission,role relationship, and information acquisition capability.

An effectivemethod tominimize the threats is to evaluatethe trusts of the interactive entities. As the main securityelement in security situation awareness mechanism [35],trust has long played a critical role in trust model or mech-anism that can suppress misbehaviors effectively in networksecurity area. Whereas, if mechanisms can process andmanage entities dynamically, as the network environmentchanges, they will evolve to situation awareness mechanismsbased on dynamic trust [68]. The situation awarenessmechanism based on dynamic trust endows the entities withthe capabilities of trust acquisition and perception [9] to assistin evaluating performance [10] andprocess real-timedecision[11], against the uncertainty, transitivity, and time decay. Thisis the principal difference between security situation aware-ness mechanism and classical network security researches,which just uses access control, authentication, and firewall,not forming a real-time decision system.

At present, there are some aspects needing to be furtherimproved in most of the existent trust mechanisms [12].(i) In time scale, the weights of historical trusts shouldconform to the regular that the more recent the trusts are,


http://dx.doi.org/10.1155/2015/107921


the higher their weights are. Furthermore, in space scale,during trust computation, many mechanisms allocate fixedweights to indirect trusts and direct trusts, not changingwith the working condition dynamically. Additionally, whentaking a third party recommender into consideration, thesemechanisms have neglected whether it has been captured ornot, thus provoking hidden trouble. (ii) Most of them haveconsidered how to obtain and compute trust without decisionsupport and risk assessment. (iii) They have held uniformtrust standard about the whole system, however, independenttrust standards are not allowed to exist in individuals, whichcannot meet the requirements of independent individuals.(iv) When confronted with some anomalous attacks orbehaviors, emergency strategy is invalid. (v) Fewmechanismshave adopted the load balance strategy, but preferring toselect the most authentic entity as the service object, whichcan bring about excessive load and resource congestion. (vi)If the malicious and eliminated entities were to attempt toregain system, time interval should have been set. (vii) In caseof the fact that entities are honest to a special service, butcheat others, different services should have different trusts ina system; however, most of mechanisms have only one typeof trust.

Aiming at these research problems, this paper proposesa decision-aided situation awareness mechanism based onmultiscale dynamic trust named DynamicTrust in wirelessnetwork, which can visualize abrupt changes about mis-behaviors. It takes social network as the research object.DynamicTrust computes integrated trust and exerts fuzzytheory to descript entity relationship and trust, consultingsuch parameters including satisfaction, indirect trust, directtrust, historical trust, and individual relevance. Buildingentities trusts based on different service types can deal withthe misbehaviors, which are honest to some service types,but cheat others. The usability, capability, vulnerability, trustauthenticity test, and trust logicality test have been usedto feed back to trust computation in period of decisionmaking and assessment. The utilization of load balanceavoids resource congestion. If we newly set environmentalparameters, this mechanism can also act on P2P, Ad hoc, andWSN, etc.

For example, if network nodes in WSN or Ad hoc shouldhave adequate sources and store capacities, we could applythe proposed system in WSN or Ad hoc, with definitionparameters in system, such as satisfaction and trust. Thenetwork nodes in WSN or Ad hoc will become the entitiesof DynamicTrust and also will collect trust information inprocess of data interaction with other nodes, computing theparameters required by the proposed system. The parametercollection and computation can be descripted as Section 3.

The remaining part of the paper is organized as follows.Section 2 mainly reviews the related works. In Section 3, weintroduce our model about situation awareness mechanismbased on multiscale dynamic trust and also present how oursituation awareness mechanism has carried out. In Section 4,we research how our model resists the confronted threatsand compare our mechanism with others by comparativelystudying simulations and performances. Finally, we concludethe paper in Section 5.

2. Related Works

These security trust models or mechanisms based on trust[616] contain the steps of information acquisition, trustcomputation, entity selection, and behavior bonus-penalty.They have considered direct trust and indirect trust, butminority of them has added feedback empiric values in trustcomputation process, andmost of them adopt recommendedtrust from a third party without checking its usability. In fact,they all cannot shape a real-time security situation awarenesssystem, not visualizing the attacks or evils and also notmaking management strategy for misbehaviors.

These existing mechanisms based on trust, for such net-works including multiagent system, P2P, MANET andWSN,have utilized Bayesian [12], expertise [13], fuzzy theory [14],evidence theory [15], bioinspired [16] and social networkgraph theory [11] to build and describe dynamic trust rela-tionship. The prestandardization of trust and reputationmodels in [17] concludes that the research based on fuzzytheory is relatively more than other researches. The existingresearches related to network security mechanisms basedon trust can be roughly divided into 4 stages from theperspective of processing sequence [18]. Originally, the firststage is information acquisition. In this stage, the mechanismcollects information, such as entity parameters, empiricalvalues, and service times. Then, the second stage is trustcomputation. The transaction weights, direct trust, indirecttrust, and empirical values are utilized to compute trustand other parameters. Moreover, the third stage is entityselection. Several rules and methods service to entity selec-tion and decision support. The last stage is bonus-penalty.Misbehaviors are doomed to punished, but honest onesneed to be rewarded. The goal of constructing the situationawareness mechanism based on dynamic trust is to providemore reliable service, make full use of system resources, andachieve profit maximization.

PeerTrust in [4] is a peer to peer communication model,whose trust lies on feedback satisfaction, service number,feedback trust, community factor, and transaction factor.PeerTrust can recognize misbehaviors and also can distin-guish false information and honest information.Whereas, allthe parameters of service providers will be used to computetrust, without considering historical trust and time relationfactor, which is the weakness of PeerTrust. Moreover, no loadbalance strategy has been applied.

A dynamic trust computation model for secured com-munication in [7], named SecuredTrust, has computed trustdepending on trust similarity, trust difference, feedback trust,and historical trust. It thinks over the time near-far effectand ponders behavior bonus-penalty. The load balancing hasalso been considered.However, only one historical trust valuebefore recent trust has been adopted, but historical trustvalues on other moments are abandoned. After misbehaviorshave been found out, no strategy has been taken.

Decision making matters in [8], named DecisionTrust,has referred usability as trust assessment parameter in wholemodel and also introduced 4 decision makings into trustmodel. Nevertheless, trust computation relies on direct trust,

International Journal of Distributed Sensor Networks 3

Integratedsatisfaction

Temporal satisfaction

Spatial satisfactionTrust

Trust type

Memorability

Trustdecision Decision 1

Level 1 trust acquisition

Level 4performance

assessment andmanagement

Level 3 decision support

Vulnerability

Usability

Level 2 trust comprehension

Capability

True

Decision 2

Opposite

Decision 3

Overstate

Decision 4

Understate

Decision 5

Collusion

Decision 6

Other

Load balance

Trust authenticity testTrust logicality test Symmetry

Transitivity

Figure 1: The decision-aided situation awareness model based on multiscale dynamic trust.

recommended trust, and environmental factor, without pon-dering on the time decay. In addition, the capability of eachentity is a given and fixed constant.

The multistrategy trust evolution model in [10] has usedfuzzy theory to compute the uncertainty and fuzziness oftrust, solving the problem of only obtaining but never sharinginformation.The game evolvementmethod has been adoptedduring trust computation, but no excessive load strategy hasbeen taken into account.

To some extent, the existing models or mechanisms canrecognize misbehaviors and improve accuracy of mechanismto actualize security communication. During trust compu-tation, historical trust has been used. However, they allcannot shape a real-time security situation awareness system,without making real-time strategies for attacks.

DynamicTrust proposed in this paper has built a deci-sion-aided situation awarenessmechanismbased on dynamictrust in wireless network. From time perspective, Dynam-icTrust computes trust and satisfaction, referring to histor-ical values and Ebbinghaus forgetting factor. From spaceperspective, after time perspective treatment, DynamicTrustwill compute parameters based on social network model andspatial relationships among entities. Moreover, Trust authen-ticity test and logicality test are used to detect the reliabilityof entities. Ultimately, the results of decision making andassessment can feed back to trust computation.

3. The Decision-Aided SituationAwareness Mechanism Based onMultiscale Dynamic Trust

3.1. SystemModel. The intent of our mechanism is to providean effective dynamic situation awareness mechanism to resistand minimize the threats. The model of DynamicTrustproposed in this paper can be divided into 4 levels, trustacquisition level, trust comprehension level, decision supportlevel, and performance assessment and management level.In first level, DynamicTrust obtains trust based on indirectsatisfaction and direct satisfaction from the scales of time and

space. In second level, DynamicTrust uses trust authenticitytest and trust logicality test to see whether the trusts providedby entities are reliable. The trust logicality contains transitiv-ity, symmetry, andmemorability. In third level, after these twotrust tests, all trusts of entities will be separated into 6 types,such as true type, opposite type, overstated type, understatedtype, collusion type, and other type. For each type, thismechanism has made a homologous strategy. In fourthlevel, DynamicTrust will manage capability, vulnerability,usability, and loads of entities, providing feedback to other3 levels and achieving the dynamic adaptation. The model ofDynamicTrust is given in Figure 1.

3.2. Trust Acquisition. This section describes several defini-tions at first level of DynamicTrust. The range of satisfactiondefinition is from 0 to 1. If entity is entirely satisfied withentity , the satisfaction of entity to entity will be 1.Otherwise, if entity is not satisfied with entity , thesatisfactionwill be 0. If an entity is incompletely satisfied withanother one, the satisfactionwill be a value in the range of 0 to1, fuzzily. Similarly, the ranges of vulnerability, capability andtrust are similar. The parameters of entities not participatingin services will keep their own values.

3.2.1. Satisfaction

Definition 1 (indirect satisfaction named SAT). Indirect sat-isfaction of entity to entity for service Sat time in th state, sat

,(, ), represents the adjacent

degree between the service satisfaction of entity to entity and the expected satisfaction of entity . We define indirectsatisfaction as sat

,(, ) SAT,

sat,

(, ) =

{{{{

{{{{

{

0, dissatisfaction,

(0, 1) , fuzzy,

1, satisfaction,

(1)


0 100 200 300 400 500 600 700 800

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0Ebbinghaus retention curve

Rete

ntio

n ca

pabi

lity

Time (/10 minutes)

Figure 2: Ebbinghaus retention curve.

where I represents th community and I is the com-munity gather of all communities. is the discrete time set ofall services, butU is the state set of all service entities. obeysthe distribution (, 1) with the cumulative distributionfunction : / = 1/2 exp{( )2/2}, where is the expected value.

Definition 2 (direct satisfaction named SAT). Direct satisfac-tion of entity for service S at time in th statecan be expressed as sat

,() = sat

,(, ) = 1, sat

,() SAT.

Definition 3 (temporal satisfaction named SAT tim). Tem-poral satisfaction of entity to entity for service S at current time in th state represents the geometricmean of all indirect satisfactions of entity at current andprevious time. We define it as sat tim

,(, ) SAT tim,

sat tim,

(, ) =

1

=0

sat

,(, ) , (2)

where represents the number of time windows, and theweight = /

=1 is proportional to , the Ebbinghaus

retention function value at time .We set0 = 1

=1,

sat00,

(, ) = 0.5, and suppose the Ebbinghaus retention func-tion value at time is 1. The Ebbinghaus retention curve [17]is like Figure 2.

Example 4. If is 4, the weights = /

=1, = 0, 1,

2, 3 of different average service response intervals will havedifferent values. The detailed weights are given in Table 1.

Definition 5 (correlated consistency named R). Correlatedconsistency between entity and entity for service S at time in th state represents the satisfaction

Table 1: Time weights based on Ebbinghaus retention rate.

Average intervalfor each service(/minute)

0 1 2 3

Interval = 10 0.342 0.271 0.200 0.187

Interval = 20 0.394 0.230 0.202 0.174

Interval = 30 0.422 0.231 0.186 0.161

correlation between arbitrary two entities , . It can beshown as () R,

() = 1

2

( 1)

1=1,2=1

12

sat tim

,(, 1) sat tim

,(, 2)

+ sat tim,

(, 1) sat tim,

(, 2)

,

(3)

where is the number of th community members. (), ranges from 0 to 1 and holds the symmetric reflex,() = ().

Definition 6 (spatial satisfaction named SAT spac). Spatialsatisfaction of entity to entity for service Sat time in th state represents the weighted mean of alltemporal satisfactions of entity at time . We defineit as sat spac

,() SAT spac,

sat spac,

() =

1

=1, = ()

=1

=

() sat tim

,(, ) ,

() > ,

(4)

where is the number of th communitymembersmeetingthe condition () > , usability > 0.5, usability is theusability of entity , and is the threshold of correlated consis-tency.

Example 7. With the known condition that there is only onecommunity E = {1, 2, 3} with = 4 and interval = 10for only one service type S, the time weights basedon Ebbinghaus retention rate will be 0 = 0.342, 1 =0.271, 2 = 0.200, and 3 = 0.187. Without considering


the existence of attacks, we suppose the indirect satisfactionsat time , 1, 2, 3 are, respectively, as follows:

SAT,

= (

0.800 0.800 0.800

0.500 0.500 0.200

0.200 0.200 0.200

) ,

SAT11,

= (

0.600 0.600 0.600

0.500 0.500 0.800

0.400 0.400 0.400

) ,

SAT22,

= (

0.400 0.400 0.400

0.500 0.500 1.000

0.600 0.600 0.600

) ,

SAT33,

= (

0.200 0.200 0.200

0.500 0.500 0.800

0.800 0.800 0.800

) .

(5)

According to the indirect satisfactions, we can get tempo-ral satisfactions

SAT tim,

= (

0.554 0.554 0.554

0.500 0.500 0.635

0.446 0.446 0.446

) . (6)

Furthermore, correlated consistency and spatial satisfac-tion are as follows:

R () = (1.000 0.910 1.000

0.910 1.000 0.910

1.000 0.910 1.000

) ,

SAT spac,

= (

0.955

0.910

0.955

) .

(7)

Definition 8 (integrated satisfaction named SAT). Integratedsatisfaction of entity at time in th staterepresents the weighted mean of spatial satisfaction anddirect satisfaction, which can be expressed as sat

() SAT,

sat

() = sat spac

,() + (1 ) sat

,() , (8)

where [0, 1] is the indirect satisfaction factor.

3.2.2. Trust Definition

Definition 9 (entity trust named T). Entity trust of entity for service S at time in th state represents theweighted mean of spatial satisfactions of all entities in . Wedescribe it as

T,

[0, 1],

=

1

=1

=1

sat spac

,() , = 1, . . . , , (9)

where is the number of th community members.

Definition 10 (local trust named TI). Local trust of commu-nity in th state represents the trust expectation of allentities in community . It also ranges from 0 to 1, describedas TI

TI,

TI

=

1

=1

, = 1, . . . , , (10)

where is the total number of communities, and TIis the

local trust of th community.

Definition 11 (globe trust named TA). Globe trust in thstate represents the whole trust of this system, which can bedescribed as

TA = 1

=1

TI. (11)

In Example 7, the trust of community E is T =(0.970 0.938 0.970)

, and the local trust is TI = (TA) =(0.959).

3.3. Trust Comprehension. Trust comprehension is at secondlevel of DynamicTrust.The main purpose of this section is tovalidate the authenticity and logicality of entity trust. Thereare two trust tests, trust authenticity test and trust logicalitytest.

3.3.1. Trust Authenticity Test. The trust authenticity test isto detect the authenticity of entity. We define the objectcollection as G in authenticity test and quantize allentity trusts into 10 grades by spacing 0.1. For arbitrary entity, according to the difference between the real trust and themean trust of entity G at th state, we can get the trustmeasure function

(G) = 1size (G)

(

1

size (G)

G

G

G

) , (12)

where size(G) is the number of the object set G elements intrust authenticity test.

is the trust of entity said by entity

, and is the real trust of entity .

If the test result is (G) = 0, DynamicTrust will considerall entities authentic in set G. However, if the consequence is(G) = 0, we consider not all entities authentic in G.

During the treat processing, the under test set D collectsthe entities being put out of G, G D = . If we regard asthe test object, the processing can be described as follows.

Step 1. Initially, G = , if the function value is (G) = 0, wecan see that all entities in G = are authentic and then dropout of the trust authenticity test. Otherwise, (G) = 0, thismechanism will get into next step.

Step 2. If the trusts of entities are equal to 0.1, we will putthem out of setG and then put setD under test. After that, wewill also compute trust measure function value again. If thevalue is (G) = 0, we will drop out of the trust authenticity


Input: All trusts of entities and

, , all mutual trusts of entities in group

Output: the collectionD in which the entities may be untruthful(1) let G = (2) if (G) = 0(3) all the entities in group is true(4) else(5) while (G) = 0(6) let = 0(7) for each do(8) if the trust of entity p is equal to 0.1(9) let get out of collection G and get into new collection D(10) end if(11) if (G) is equal to 0(12) break(13) end if(14) end for(15) end while(16) end if

Algorithm 1: Truth detection of entities in group (

,

, , ).

test. In contrast, if the trusts of entities are equal to 0.2, wewill put them out of set G and then also put D under testset, with repetitive operation and computation. According tothe sequence of trusts from low to high, we will repeat theoperation sequentially, until we get (G) = 0 or the set G isempty.

The algorithm in trust authenticity test is given inAlgorithm 1.

3.3.2. Trust Logicality Test

Definition 12 (trust difference named ). The trust differencebetween the real trust of entity and the trust of entity saidby entity at th state, which can be described as = |

|, . If entity trusts entity , we will set = 0.

The under test set D is the object collection in trustlogicality test. After trust authentic test, if under set D is notempty, we will carry out trust logical test for entities in D.There are three aspects in trust logicality test, such as transi-tivity test, symmetric consistency test, and memorability test.

(a) Symmetric Consistency Test. The trust of entity said byentity should be equal to that of entity said by entity ,

=

, or else we will take , , D as unbelievable

entities which will be doomed to enter next stage decisionsupport.

(b) Transitivity Test. In view of the fact that multilevel trans-mission may bring up the expending of attacks, we settransitivity with only one level. For example, if entity trustsentity and entity trusts entity and entity trusts entity,we can know entity trusts entity but do not knowwhetherentity trusts entity .

Theorem 13. With the known condition = = 0, if ,, are credible, we can get = 0.

Proof. From = |

| = 0 and = |

| = 0, we

can see = |

| = |

| = 0. In other words,

if trusts and trusts , will trust . However, if thereare = = 0 and = 0, , , will be considered asincredible, which is opposite to the known condition. There-fore, = 0.

(c) Memorability Test. Compared with historical entity trust,if current entity trust is higher or identical, we will think theentity is credible. However, if current entity trust is lower, wewill consider it as an unbelievable entity.

Those entities not passing the trust authenticity test ortrust logicality test, which will be regarded as unbelievableentities, will come under decision set and enter next stage,decision support.

3.4. Decision Support. Decision support exists at third level ofDynamicTrust. After trust acquisition and trust comprehen-sion, in this section, DynamicTrust will determine the trusttype and then make decision for different trust types.

3.4.1. Trust Type Decision

Definition 14 (trust deviation named ). The trust deviationbetween the real trust and themean trust of entity in at thstate represents the mean difference between the real trust ofentity and the trusts of entity said by other entities in, which can be described as () ,

() =

1

1

=1, =

(

) , (13)


where is the actual trust of entity said by entity and

is the real trust of entity .

Taking the possible contingencies into consideration, weclassify trusts into 6 types, such as true type, opposite type,overstated type, understated type, collusion type, and othertype.

True Type. If there is () = 0, entity will be considered ashonest. We take such entity into true type.

Opposite Type. If there are () = 0 and

+

= 1, trust of

entity will be considered as opposite, where is the mean

trust of entity agreed by all entities in . We put entity intoopposite type.

Overstated Type. If there is () > 0, trust of entity will beoverstated. We take entity into overstated type.

Understated Type. If there is () < 0, trust of entity will beunderstated. We put entity as understated type.

Collusion Type. If there are more than 3 entities whose trustsare identically opposite or overstated or understated in a samecommunity, we will take such entities into collusion entities.

Other Type. If entities do not belong to the former 5 types, wewill collect them into other types.

3.4.2. Trust Decision Support

Definition 15 (deviation factor named ). The deviationfactor of the trust deviation to the real trust of entity in th state represents the absolute value of the ratio of ()to , which can be described as ,

=

()

, 0 1. (14)

Decision 1. For entity of true type, we will adapt its trustas

=

.

Decision 2. For entity of opposite type, we will modifyits trust as

=

.

Decision 3. For entity of overstated type, we will adjustits trust as

= (1 )

, 0 1.

Decision 4. For entity of understated type, wewillmodifyits trust as

= (1 )

, 0 1.

Decision 5. For entities of collusion type, wewill not onlyadapt trust based on its corresponding trust type, but will alsoreduce trust to

=

.

Decision 6. To entity of other types, we allocate a valuein the range of [0, 1] to the trust, randomly.

This system will filtrate out entities with trust not in therange of [0.2, 1]. After a period of time, they can be allowedto reenter the system. This mechanism will also redistributeinitial parameters for all entities, after executing 50 services.

3.5. Aided Performance Assessment. This section is at fourthlevel of DynamicTrust. To cope with the condition that anentity with high trust is compromised at current service, butwe also look on it as reliable entity, we introduce usability intothis mechanism, to reduce the probability of attack success.We take integrated trust as the capability.

Definition 16 (entity capability named CAP). Entity capabil-ity of entity at time in th state represents theintegrated satisfaction of entity , which can be expressedas cap

() CAP,

cap

() = sat

() . (15)

If a new entity enters the system, system will initializeits capability as cap0

0() = 0.5.

Definition 17 (community capability named CAPI). Com-munity capability of I at time in th state representsthe expectation of all capabilities of entities in , which canbe expressed as cap

CAPI,

cap =

1

=1

cap

() . (16)

Definition 18 (entity vulnerability named V). Entity vulnera-bility of entity at time in th state represents therecuperative capability, which can be expressed as vul

()

V,

vul

() = 1 cap

() . (17)

Definition 19 (entity relative usability named usability).Entity relative usability of entity at time in th staterepresents the ratio of entity capability to the maximumcapability in at current state, which can be expressed asusability(, ) usability, U CAP usability,

usability (, ) =cap

()

max cap

()

() , ,

(18)

where () U,

() {0, 1} is the state of entity at time

in th state. () = 0 represents the entity which

is unavailable, but () = 1 represents entity which is avail-

able. If a new entity enters the system, system will initializeits usability as usability(, 0) = 1.

Supposing that the capability of entity is cap() = 0.8

and the maximum capability is maxcap

() = 1, if the

entity is in the system at time , () = 1, its usability can

be expressed as usability(, ) = 0.8. In contrast, if it is not inthe system, its usability will be usability(, ) = 0.

Example 20. In Example 7, with the known parameters,SAT spac

,= (0.955, 0.910, 0.955)

, sat,

(1) = 0.554,


Table 2: Service parameter table.

Service type Significant degree The parameters of entities providing serviceLogicality Trust Usability Remarka

Type Most significant Satisfaction

> 1 Usability > 0.8 Selecting the entity providing service for type Type Special significant Dissatisfaction

> 2 Usability > 0.6 Selecting the entity providing service for type

Type Significant Dissatisfaction

> 3 Usability > 0.5 Selecting the entity providing service for type Type General Dissatisfaction

> 4

Usability > 0.4 Selecting service entity, randomlyaWhat should we select to provide service, if there is no entity meeting the parameter requirement?

sat,

(2) = 0.500, sat,

(3) = 0.446, and = 0.7, we can obtainthe entity capabilityCAP

= (0.835, 0.787, 0.802)

, the entityvulnerability V

= (0.165, 0.213, 0.198)

, and the entity rela-tive usability usability

= (1.000, 0.943, 0.960)

.

3.6. Load Balance Related to Select Service Object. Thissection introduces a load balance method at fourth levelof DynamicTrust. According to the significance of service,services have been separated into 4 types, such as type ,type , type , and type , thus constituting service set S ={, , , }. Different service objects are used for differentservice type.The trust threshold set can be = {1, 2, 3, 4}.The service parameters are as Table 2.

In load balance, the related steps are listed as follows.

Step 1. After entity makes a request for service , all entitiesresponding to the request compose the response set . Wedefine the response entity finally providing service as entity . , if entity passes the trust logicality test andmeets thecondition usability(, ) > ,

> , we will bring it into the

credible set . However, if entity only meets the conditionusability(, ) > ,

> , but does not satisfy the trust

logicality, it will be put into the candidate set. All the entitiesin but not in or will be put into the second choiceset .

Step 2. If set is nonempty, , we will compute the loadof entity , (). Then, we will find out entities with smallestloads. Finally, we randomly select an entity as entity from entities.

Step 3. If credible set is empty but the candidate set isnonempty, , we will compute the load of entity , (),selecting the entity with smallest load as entity .

Step 4. If sets and are empty, wewill select response entityfrom = . , if entity meets the condition

> 0.5,

we will compute the load of entity , (), and select the entitywith smallest load as entity . Otherwise, we will randomlyselect response entity from .

Assuming that is 1, the load balance is given asAlgorithm 2.

4. Simulation Analysis andPerformance Comparison

In this section, we simulate the decision-aided situationawareness mechanism based on multiscale dynamic trustrelaying on the above theoretical frame to evaluate the mech-anismperformance and prove the applicability and effectivity.We have fulfilled our simulation at MATLAB 7.1 simulationplatform inWindows operating system with Intel Core (TM)Duo 2.66GHz CPU, 2GB Memory. There are entities andthe total number of service times is in our simulation. Thelength of time window is . We set as 100, classifying allservices into 4 communities 1, 2, 3, 4, with 27, 25, 22, 26entities, respectively. We can get I = 1 2 3 4. Thesimulation regards community 3 as the research object. Theparameter setting is given in Table 3.

In this part, we compare our DynamicTrust with Secur-edTrust [7], PeerTrust [4] and DecisionTrust [8], from 4aspects, including sensitivity and consistency evaluation, sta-bility evaluation, usability evaluation and load balance evalu-ation.

4.1. Sensitivity and Consistency Evaluation

Definition 21 (sensitivity named sensitikity). The sensitivityof entity in th state represents the average deviation degreebetween the real trust and the actual trust of entity , whichcan be described as sensitivity

sensitikity,

sensitivity

=

max (, )

. (19)

Based on above theory, trust is a significant parameter intrust model. In community 3, we artificially set the frequen-cies of attacks at 2nd entity, 4th entity, 6th entity, 10th entityas ma res = 0%, ma res = 12.5%, ma res = 25%, ma res =100%, respectively. If the real trusts of these 4 entitiesare 0, the attacks situation distribution will be like Figure 3.

As is shown in Figure 3, 4th entity will tell other entitiesthat its own real trust is 0.6 every 4 services, but 6th entitywill tell other entities that its own real trust is 0.6 every 8services. The 10th entity will always report its trust value as0.6, maliciously. The 2nd entity will always report the realtrust value. After we compare the trusts and sensitivities of


Input: Entity , all trusts , and all usability(, ), of entities

responding to for service , and under decision set in which elements are illogic.Output: Entity (1) for each do(2) if usability(, ) > , > and then(3) put into collection (4) else if usability(, ) > , > and then(5) put into collection (6) else(7) put into collection (8) end if(9) end if(10) end for(11) if = then(12) for each do(13) compute the load ()(14) return the entity with the smallest load(15) end for(16) else if = then(17) for each do(18) compute the load ()(19) return the entity with the smallest load(20) end for(21) else(22) for each do(23) if

> 0.5 then

(24) compute the load ()(25) return the entity with the smallest load(26) else(27) return the entity , randomly(28) end if(29) end for(30) end if(31) end if

Algorithm 2: Load balance for entity (, usability(, ), , ).

these 4 entities according to the known condition in Figure 3,we will gain the situation comparison as Figure 4.

In Figure 4, we can see the sensitivity of SecuredTrust isthe lowest. The sensitivities of other 3 entities are relativelybigger. Most of trusts in these four models are from 0.5 to 1.In these 4 sub-figures, only in DecisionTrust model can the2nd entity trust reach 1, but cannot reach 1 in other 3 models,because other models have used community parameters tocompute trust. If there is a malicious entity, all entity trustscannot reach 1, but can be extremely near to 1.

In Figure 4(a), if there is no attack, the entity trust willrise, but when there exists an attack, it will decline very soon.The 10th entity always launching attacks has been eliminatedandnever returned the system, after providing the 8th service,which may mean DecisionTrust has overestimated the effectsof attacks.

In Figure 4(b), if there is no attack, the entity trust willkeep, but when there exists an attack, it will decline with rel-ative smaller amplitude. The 6th entity trust should be lowerthan the 4th entity trust and the 10th entity trust, factually.

However, some trust values of 4th entity are higher than 6thentitys, not agreeing with the fact.

In Figure 4(c), if there is no attack, the entity trust willkeep and slowly go up, but when there is an attack, it willdeclinewith smaller amplitude.Whereas, nomatter the entityis malicious, the entity trusts are always from 0.55 to 0.85, toa large extent, which reveals the extreme underestimation ofthe whole system.

In Figure 4(d), if there is no attack, the entity trust willretain and slowly rise, but when there is an attack, it willdecline with relative smaller amplitude. There is no overesti-mation and underestimation in DynamicTrust model, whichis relatively eclectic and realistic.

4.2. Stability Evaluation. The mechanism regards the sen-sitivity variance named SV as norm to evaluate stability,referring to [19].

Definition 22 (sensitivity variance named SV). The sensitivityvariance of entity represents the fluctuation of the average


Table 3: Parameter settings.

Settings Parameters Description Default

Simulation environment settings

#The total number of entities in system 100

3

#The total number of community 3 insystem 22

#The number of malicious entities 0%

ma res #The frequency of a malicious entitylaunching attack 0%

ma rep #The percentage of collusion attack inmalicious attacks 0%

interval #The average interval for each service(/minute) 10

#The total number of service times 60

Parameter computation settings

#The trust threshold about entities {0.8, 0.6, 0.5, 0.2} #The indirect satisfaction factor 0.3 #The length of time window 4

#The expectation value of each service incomputation indirect satisfaction 1

#The output number of function min() inload balance 1

#The threshold of entity usability in loadbalance 0.5

#The threshold of entity trust in load balance 0.4

Table 4: Stability comparison.

SV DynamicTrust SecuredTrust PeerTrust DecisionTrustSV of the 2nd entity 0.0000 0.0000 0.0000 0.0000SV of the 4th entity 0.0072 0.0072 0.0016 0.0165SV of the 6th entity 0.0141 0.0141 0.0015 0.0100SV of the 10th entity 0.0513 0.0513 0.0003 0.0005

0.4010 20 30 40 50 60

0.45

0.50

0.55

0.60

0.65

Service times

The r

eal t

rust

valu

e

Attacks situation

Attacks situation of the 2nd entity Attacks situation of the 4th entity Attacks situation of the 6th entity Attacks situation of the 10th entity

0

Figure 3: Attaks distribution situation.

deviation degree between the real trust and the factual trustof entity , which can be described as

SV =1

(mean1

[sensitivity] sensitivity

)

2

, (20)

where is the total number of service times.

Malicious entities want to alter its own trust to misleadother entities, which will arouse the fluctuations of entitytrust. According to sensitivity variances of 2nd, 4th, 6th, and10th entities in DynamicTrust, SecuredTrust [7], PeerTrust[4], and DecisionTrust [8] models, we can obtain varianceslisted as Table 4 based on sensitivity evaluation in Figure 4.

In Table 4, we can see that themean SV of PeerTrust is thesmallest, but that of DynamicTrust and that of SecuredTrustare higher. After 60 services, the SV of 10th entity in Dynam-icTrust reaches 0.05, which is the same in DecisionTrust.Luckily, all sensitivity variances of entities in those 4 modelsare small.

4.3. Entity Relative Usability. Based on the known conditionin Figure 3, we can also gain the usability of 2nd, 4th, 6th and


Trus

t and

sens

itivi

ty

0 10 20 30 40 50 60

0.10.20.30.40.50.60.70.80.91.0

Service times

Sensitivity of 2nd entity Sensitivity of 4th entitySensitivity of 6th entitySensitivity of 10th entity

Trust of 2nd entityTrust of 4th entity Trust of 6th entity Trust of 10th entity

(a) DecisionTrust trust and sensitivity situation

Sensitivity of 2nd entity

Trus

t and

sens

itivi

ty

0 10 20 30 40 50 60

0.10.20.30.40.50.60.70.80.91.0

Service times

Sensitivity of 4th entitySensitivity of 6th entitySensitivity of 10th entity

Trust of 2nd entityTrust of 4th entity Trust of 6th entity Trust of 10th entity

(b) PeerTrust trust and sensitivity situation

10 20 30 40 50 60Service times


Trust of 2nd entity Trust of 4th entity Trust of 6th entity Trust of 10th entity

0

0.10.20.30.40.50.60.70.80.91.0

Trus

t and

sens

itivi

ty

(c) SecuredTrust trust and sensitivity situation


Trust of 2nd entity Trust of 4th entity Trust of 6th entity Trust of 10th entity

0 10 20 30 40 50 60Service times

0.10.20.30.40.50.60.70.80.91.0

Trus

t and

sens

itivi

ty

(d) DynamicTrust trust and sensitivity situation

Figure 4: Trust and sensitivity situation.

10th entities as Figure 5. There is a regular that the usabilityof 2nd entity should be higher than that of 4th entity and thatof 6th entity.The usability of 10th entity should be the lowest.

From Figure 5(a), we know most of the relative usabilityof entities is from 0.6 to 1 and they meet the usabilityregular. However, the 10th entity has also been eliminated andnever returned the system, after providing the 8th service. InFigure 5(b), the relative usability is from 0.75 to 1, but doesnotmeet the usability regular. Sometimes, the usability of 10thentity is higher than that of 4th entity, not in line with the fact.In Figures 5(c) and 5(d), the usability of entities both meetsthe usability regular and is in line with reality.

4.4. Load Evaluation. In this section, we suppose there are1000 services and 12 entities in a system. The 1000 services

contains 250 type services, 250 type services, 250 type services and 250 type services. The trust threshold setis = {0.8, 0.7, 0.5, 0.4}. The entity parameters are listed asTable 5.

We suppose that entity trust threshold and usabilitythreshold are both 0.7. After an entity emits service request,PeerTrust will randomly select service object from the mostcredible entities with trusts bigger than 0.7. DecisionTrustwill always select service object from the entities whosetrusts and usability are both bigger than 0.7. SecuredTrusthas its own load balance strategy, which preferentially selectsservice object from the entities with trusts bigger than 0.7or randomly selecting entity as service object. DynamicTrustwill select service object, according to the service type, load,logicality, usability, and entity trust. For type services,


0 10 20 30 40 50 60Service times

Usa

bilit

y

Usability of 2nd entityUsability of 4th entity

Usability of 6th entityUsability of 10th entity

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

(a) DecisionTrust usability



0 10 20 30 40 50 60

Usa

bilit

y

0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Service times

(b) PeerTrust usability

0 10 20 30 40 50 60



0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Service times

Usa

bilit

y

(c) SecuredTrust usability

0 10 20 30 40 50 60



0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Service times

Usa

bilit

y

(d) DynamicTrust usability

Figure 5: Entity relative usability.

Table 5: Entity parameter settings.

Entity 1 2 3 4 5 6 7 8 9 10 11 12Entity trust 0.95 0.96 0.88 0.85 0.85 0.75 0.71 0.6 0.55 0.5 0.5 0.45Logical Yes Yes Yes No No No No No No No No NoUsability 0.85 0.88 0.8 0.79 0.78 0.72 0.68 0.65 0.6 0.54 0.66 0.5

the entities with higher usability and trusts can be serviceobjects. For type services, the entities with lower usabilityand trusts can become service objects, thusmaking full use ofall entities existing systems.We have obtained Figure 6, com-paring the loads of DynamicTrust, PeerTrust, DecisionTrust,and SecuredTrust.

As is indicated in Figure 6, PeerTrust will always select1st entity, or 2nd entity, or 3rd entity as service object.

The average load of PeerTrust is high. DecisionTrust willselect service object from 1st entity to 6th entity, so its averageload is lower. SecuredTrust will select service object from 1stentity to 7th entity. DynamicTrust will select service objectwith load balance strategy for different services. For type servicewith lower requirement, DynamicTrust selects serviceobject from 8th entity to 12th entity. The average load ofDynamicTrust is the lowest.


1 2 3 4 5 6 7 8 9 10 11 120

50

100

150

200

250

300

350

400

Entity ID

Load

DynamicTrustSecuredTrust

PeerTrustDecisionTrust

Figure 6: Load comparison.

5. Conclusion

Aiming at existing research problems, this paper proposesa decision-aided situation awareness mechanism based onmultiscale dynamic trust in wireless network. DynamicTrustcomputes and defines satisfaction, trust, and other param-eters based on Ebbinghaus forgetting regular from timeperspective and spatial relationships from space perspective.We have also used usability, capability, and trust tests to formfeedback and aid decisionmaking and assessment. Comparedwith 3 other models, DynamicTrust is relatively eclectic andrealistic for trust mensuration, which can also make full useof entities in system, avoiding resource congestion. However,the trust situation may arise interrupted. To emergencies,the mechanism should make more perfect strategies basedon historical and current information, which needs a large-scale database. That is the challenge of situation awarenesstechnology, remaining to be improved.

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper.

Acknowledgments

This work is supported by the National Nature ScienceFoundation of China (nos. 61271260; 61102062; 61301122),NSF of Chongqing (no. cstc2014jcyjA40052), the ResearchProgram of Chongqing Municipal Education Commission(no. KJ1400405), Program for Changjiang Scholars and Inno-vative Research Team in University (IRT1299), the specialfund of Chongqing key laboratory (CSTC), NSF of CQUPT(no. A2013-30), and the Doctor Science Research StartingFoundation of CQUPT (no. A2013-23).

References

[1] J. Webb, A. Ahmad, S. B. Maynard, and G. Shanks, A situationawareness model for information security risk management,Computers & Security, vol. 44, pp. 115, 2014.

[2] K. R. Sarkar, Assessing insider threats to information securityusing technical, behavioural and organisational measures,Information Security Technical Report, vol. 15, no. 3, pp. 112133,2010.

[3] S. Ganeriwal, L. K. Balzano, and M. B. Srivastava, Reputation-based framework for high integrity sensor networks, ACMTransactions on Sensor Networks, vol. 4, no. 3, article 15, 37pages, 2008.

[4] L. Xiong and L. Liu, PeerTrust: supporting reputation-basedtrust for peer-to-peer electronic communities, IEEE Transac-tions on Knowledge and Data Engineering, vol. 16, no. 7, pp. 843857, 2004.

[5] F. L. Li, Y. F. Nie, F. Liu, J. Zhu, and H. B. Zhang, Event-centricsituation trust data aggregation mechanism in distributedwireless network, International Journal of Distributed SensorNetworks, vol. 2014, Article ID 585302, 11 pages, 2014.

[6] Y. Hou, Y. Xiong, X. Wang, and X. Liang, The effects of atrust mechanism on a dynamic supply chain network, ExpertSystems with Applications, vol. 41, no. 6, pp. 30603068, 2014.

[7] A. Das and M. M. Islam, SecuredTrust: a dynamic trust com-putation model for secured communication in multiagent sys-tems, IEEE Transactions on Dependable and Secure Computing,vol. 9, no. 2, pp. 261274, 2012.

[8] D. Jelenc, R. Hermoso, J. Sabater-Mir, and D. Trcek, Deci-sion making matters: a better way to evaluate trust models,Knowledge-Based Systems, vol. 52, pp. 147164, 2013.

[9] Y. A. Kim and H. S. Song, Strategies for predicting local trustbased on trust propagation in social networks, Knowledge-Based Systems, vol. 24, no. 8, pp. 13601371, 2011.

[10] G. Yin, Y. Wang, Y. Dong, and H. Dong, Wright-Fisher multi-strategy trust evolution model with white noise for Internet-ware,Expert Systemswith Applications, vol. 40, no. 18, pp. 73677380, 2013.

[11] J. Wu and F. Chiclana, A social network analysis trust-consensus based approach to group decision-making problemswith interval-valued fuzzy reciprocal preference relations,Knowledge-Based Systems, vol. 59, pp. 97107, 2014.

[12] X. Yang, Y. Guo, and Y. Liu, Bayesian-inference-based recom-mendation in online social networks, IEEE Transactions onParallel andDistributed Systems, vol. 24, no. 4, pp. 642651, 2013.

[13] Z.-P. Fan, W.-L. Suo, B. Feng, and Y. Liu, Trust estimation in avirtual team: a decision support method, Expert Systems withApplications, vol. 38, no. 8, pp. 1024010251, 2011.

[14] M. Brunelli and M. Fedrizzi, Fuzzy m-ary adjacency relationsin social network analysis: optimization and consensus evalua-tion, Information Fusion, vol. 17, no. 1, pp. 3645, 2014.

[15] M. W. Hassan, R. McClatchey, and I. Willers, A scalable evi-dence based self-managing framework for trust management,Electronic Notes in Theoretical Computer Science, vol. 179, pp.5973, 2007.

[16] H. Marzi and M. Li, An enhanced bio-inspired trust and repu-tation model for wireless sensor network, Procedia ComputerScience, vol. 19, pp. 11591166, 2013, Proceedings of the 4thInternational Conference on Ambient Systems, Networks andTechnologies.

[17] F. G.Marmol and G.M. Perez, Towards pre-standardization oftrust and reputation models for distributed and heterogeneous


systems, Computer Standards & Interfaces, vol. 32, no. 4, pp.185196, 2010.

[18] K. Govindan and P. Mohapatra, Trust computations and trustdynamics in mobile adhoc networks: a survey, IEEE Commu-nications Surveys and Tutorials, vol. 14, no. 2, pp. 279298, 2012.

[19] S. V. Halunga and N. Vizireanu, Performance evaluation forconventional and MMSE multiuser detection algorithms inimperfect reception conditions, Digital Signal Processing, vol.20, no. 1, pp. 166178, 2010.

Review ArticleA Survey on Multihop Ad Hoc Networks forDisaster Response Scenarios

D. G. Reina,1 M. Askalani,1 S. L. Toral,1 F. Barrero,1 E. Asimakopoulou,2 and N. Bessis2

1Electronic Engineering Department, University of Seville, Camino de los Descubrimientos S/N, 41092 Seville, Spain2School of Computing & Maths, University of Derby, Kedleston Road, Derby DE22 1GB, UK

Correspondence should be addressed to F. Barrero; [email protected]

Received 23 January 2015; Revised 20 April 2015; Accepted 11 May 2015

Academic Editor: Mauro Conti

Copyright 2015 D. G. Reina et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Disastrous events are one of the most challenging applications of multihop ad hoc networks due to possible damages of existingtelecommunication infrastructure.The deployed cellular communication infrastructure might be partially or completely destroyedafter a natural disaster. Multihop ad hoc communication is an interesting alternative to deal with the lack of communications indisaster scenarios.They have evolved since their origin, leading to different ad hoc paradigms such asMANETs, VANETs, DTNs, orWSNs. This paper presents a survey on multihop ad hoc network paradigms for disaster scenarios. It highlights their applicabilityto important tasks in disaster relief operations. More specifically, the paper reviews the main work found in the literature, whichemployed ad hoc networks in disaster scenarios. In addition, it discusses the open challenges and the future research directions foreach different ad hoc paradigm.

1. Introduction

Ad hoc networks have been proposed as an appealingcommunication technology to deal with the unexpectedconditions emerging during and/or after the occurrenceof a disaster. Communication between victims and crewmembers involved in rescue operations is crucial in order toalleviate the disaster consequences and save lives. The first72 hours after the occurrence of the disaster are the mostimportant according to some studies [1, 2]. This period oftime is called the golden relief time. After the golden relieftime, the probability of finding survivors is very low. Con-sequently, coordination of first responders and/or victims isof paramount importance. Communications in general andamong the first responders in particular play an importantrole in achieving an efficient coordination. Nowadays, peoplemostly communicate with each other by mobile phones andsmartphones, making calls or sending text messages throughInternet and social networks via applications such as What-sApp, Facebook, Twitter, and others [3]. However, cellular-based communications may not be possible after a disaster

due to the damages in the telecommunication infrastructure,leaving many people isolated and unprotected. In [4], theauthors present Disaster Recovery Networks (DRN) andSearch and Rescue Network (SRN) as the main networksneeded for disaster relief. The objective of a DRN consists ofproviding emergency support to victims and crew memberstaking part in rescue operations. On the other hand, a SRNis a network formed to track individuals in an emergencyoperation. In [4], the authors state the main features requiredfor DRN and SRN, such as quick response, life expectancy ofthe network, interoperability, tariff-free operation, networkcoverage, support for heterogeneous traffic types, networkcapacity, ease of use and cost of equipment, outdoor andindoor operation, high precision for localization, and searchoperation. Mobile ad hoc networks [4, 5] exhibit manyof the above-mentioned features so they are suitable forboth DRN and SRN and consequently for disaster responsenetworks in general. Furthermore, in [6], the authors statethe key obstacles in effective disaster response, such ascommunication and collaboration support, provision of real-time data to field personnel and to incident command post,


http://dx.doi.org/10.1155/2015/647037


unified approach to data handling, visual data capture, on-sitebuilding assessment, access to building design documents,personal mobility support, resource allocation issues, andmultiple connectivity. Ad hoc networks can be a feasiblesolution for those complications related to real-time commu-nication and collaboration between the personnel taking partin the rescue operations.

The basic idea of ad hoc networks is based on defining amultihop communication path between two or several nodesin the network.This initial idea has been constantly evolving,leading to the development of new ad hoc paradigms such asVehicular Ad Network (VANETs), Delay Tolerant Networks,Wireless SensorNetworks (WSNs) [7], and others.This paperaddresses the applicability of such multihop ad hoc networksin disaster response scenarios.Themain contributions of thispaper are listed as follows:

(i) A detailed description of the existingmultihop ad hocnetwork paradigms applicable to disaster responsescenarios, including types of communications, wire-less technologies, and research directions.

(ii) A thorough review of the existing work for each adhoc paradigm, highlighting the necessity of interop-erability among different ad hoc paradigms.

(iii) A review of the future open challenges for multihopad hoc networks in disaster response scenarios.

This paper continues as follows. Section 2 presents theevolution of multihop ad hoc networks and the different adhoc paradigms that can be used in disaster scenarios, describ-ing the main features of each paradigm. Existing works foreach ad hoc network in disaster scenarios are reviewed inSection 3. Section 4 discusses the open challenges and futureresearch directions, and finally the main conclusions of thissurvey are presented in Section 5.

2. Evolution of Ad Hoc NetworkParadigm and Its Application toDisaster Response Scenarios

In the classical viewpoint of ad hoc networks, nodes com-municate with each other without requiring a central systemor any type of infrastructure. Thus, nodes act as routers andhosts at the same time, being responsible for generating thecontent and disseminating it efficiently throughout nearbynodes in the network. Under this communication paradigm,mobility is actually a problem in classical MANETs (MobileAd hoc Networks), where routing protocols must deal withmobility of nodes and continuous topological changes inorder to establish and maintain a communication pathbetween a pair of source-destination nodes. The first idea ofMANET was the replication of wired Internet communica-tion like TCP/IP communications in mobile environments.However, in mobile scenarios, the replication of TCP/IPcommunications through a multihop path is not an easytask. The MANET paradigm has evolved since its bornleading to new ad hoc paradigms based on the basic ideaof communicating electronic devices through a multihopcommunication path and in a decentralized way.

2.1. The Evolution of Multihop Ad Hoc Networks. One of thefirst evolutions of traditional MANET is VANETs [7], whichconsider cars as the carriers of wireless transceiver, so carscan communicate with each other. The main idea behindVANETs is similar to that of MANETs but with a highermobility of nodes and being limited to traffic lanes. There arealso other clear differences in the wireless technologies used,MAC and application layers. Furthermore, Delay TolerantNetworks (DTNs) have arisen in the last decade as one ofthe most interesting evolutions of classical MANETs [52]. InDTNs, nodes do not require a high connectivity in order tocommunicate with each other. The transmitted informationis not delay-sensitive so nodes follow the carry-store and for-ward paradigm in which nodes generate certain informationand store it until they have a new opportunity to deliver it,that is, whenever they meet other nodes. For this feature,DTNs are also referred to as opportunistic networks, whereeach encounter between two nodes in the network is seenas a new opportunity to deliver information. In contrast toMANETs, mobility is seen as an advantage for disseminatinginformation inDTNs since the higher themobility, the higherthe number of possible encounters with other nodes.

Other static ad hoc networks such as WSNs [53] orWireless Mesh Networks (WMNs) [54] are also evolutionsof classical MANETs applicable to disaster scenarios. Withregard to mobility, WSNs are normally static networks wherenodes sense environmental variables such as temperature andCO2emissions. With respect to network hierarchy, in WSNs,

there is normally a central node in charge of gathering senseddata from other nodes, processing the data and sending thisinformation to awider range of networks such as the Internet.WMNs are normally applied to extend Internet connectivityto areas where traditional connections via cable are difficult,if not impossible.

2.2. Multihop Ad Hoc Communications for Disaster Scenarios.This subsection is aimed at presenting the main featuresof the different ad hoc communication paradigms suchas MANETs, VANETs, DTNs, WSNs, WMNs, RFID, andTETRA and, most importantly, their applicability for disasterscenarios. Table 1 summarizes the main features of the adhoc communication paradigms that will be described in thissection. For more information about the different ad hocnetworking paradigms and wireless technologies, the readercan refer to [52, 5557].

2.2.1. MANETs. Two types of communications can be con-sidered in classical MANETs, broadcast and multihop com-munications via routing protocols.

In broadcast communications, a node shares the sameinformation simultaneously with its one hop neighbouringnode. It may be suitable for transmitting warning messagesand alarms, which are crucial forms of communications indisaster scenarios. Although this type of communicationis the simplest way of transmitting data among nodes, forexample, employing simple flooding, it may exhibit someproblems in congested scenarios due to the broadcast stormproblem [58]. A basic classification of broadcasting schemesin ad hoc networks categorizes the schemes into two groups,


Table 1: Main features of multihop ad hoc communication paradigms.

Ad hocparadigm Mobility

Expecteddensity Wireless technology Topology Routing Main functionality

MANET Low-mediumUnexpected High

WiFi (IEEE802.11a/b/g)Bluetooth

(IEEE 802.15.1)

ChangeableFlat

Broadcasting schemesRouting protocols(Network layer)

Real-timecommunications

VANET

HighPredictable andconstrained by

lanes

Medium-highWAVE, ISO CALM

DSRC(IEEE 802.11p)

ChangeableFlat

Broadcasting schemesRouting protocol(Network layer)


DTN MediumUnexpected Medium-low

WiFi (IEEE802.11a/b/g)Bluetooth

(IEEE 802.15.1)

ChangeableFlat

Forwarding schemes(Bundle layer)

Communicationsfor nondelayedsensitive data

WSN Low High ZigBee(IEEE 802.15.4)Fixed

Hierarchical

Broadcasting schemesRouting protocols(Network layer)

Detection andwarning systems

WMN Low HighWiFi (IEEE802.11a/b/g)WiMAX

FixedHierarchical

Broadcasting schemesRouting protocols(network layer)

Backbone network

RFID Very low High NFC Hierarchical Point-to-point IdentificationTETRATETRAPOL(direct mode)

MediumTactical Medium

Private (similaritieswith GSM)

ChangeableFlat and

hierarchical

Point-to-pointPoint-to-multipoint

Repeater


namely, the deterministic and the probabilistic schemes. Indeterministic schemes, a subset of the nodes in the networksis selected as the potential forwarders to retransmit thebroadcast packets [59]. In contrast to deterministic schemes,in probabilistic protocols, nodes retransmit the broadcastpackets with a precalculated forwarding probability [60, 61].

When nodes rely on routing protocols for establishinga communication path to a destination node, nodes shouldmaintain routing tables in order to select which is the bestnext hop to route the information. Regarding the types ofrouting protocols, two main categories are distinguished:proactive routing protocols and reactive routing protocols[62].Themain difference between both types is that, in proac-tive protocols, nodes exchange routing tables and topologyinformation periodically. In contrast, in reactive protocols,routes are only created and maintained whenever they areneeded and they are active. Research on routing protocolsfor ad hoc networks has been an active field for the lasttwo decades [62, 63]. They can be applicable to importanttasks in disaster scenarios. Let us imagine a situation wherea rescue team needs to retransmit certain information to acentral unit. Amultihop route can be established from a crewmember to the central unit. The routing protocol would beresponsible for selecting the best nodes to retransmit the datapackets from the crew member to the central unit.

Regarding the wireless technology used in MANETs, theIEEE 802.11a/b/g has been the technology usually employed.The Wireless Fidelity (WiFi) standard is based on IEEE802.11 technology and it is used as the main technologyto connect smartphones to the Internet access points inhome connections. However, the applicability of IEEE 802.11

technology in MANETs requires the IEEE 802.11 wirelesstransceiver to be configured in Distributed CoordinationFunction (DCF) in order for nodes to function in ad hocmode, which is different from the normal operation of IEEE802.11 in smartphones. Bluetooth technology is also includedin smartphones and, consequently, it is a potential candi-date to be used in MANETs. However, Bluetooth requiresthe manual configuration of some parameters in order toestablish communications. Another shortcoming is that theuse of Bluetooth in smartphones implies a considerableincrease of power consumption reducing the autonomy ofthe smartphones, which is a crucial parameter in disasterscenarios. Recently, WiFi Direct has appeared as appealingtechnology for ad hoc communications. However, it is notwidely used in smartphones yet, but it must be considered innear future.

2.2.2. VANETs. Regarding the application of VANETs indisaster scenarios, it is obvious that communication amongvehicles can also play an important role in reducing theconsequences of disasters. Situations such as traffic accidents,reorganization of traffic by reducing congestion throughproposing alternative routes, localization of victims, andcommunications between ambulances and victims or otherrescue teams like firefighters or police officers are only fewpossible applications of the VANET paradigm in disasterscenarios. The basic types of communications in VANETsare the same above-mentioned in MANETs, so nodes (vehi-cles) can use both broadcast communications and multihopcommunications via routing protocols [64]. However, thereare significant differences in the design of broadcasting


and routing protocols for VANETs. Firstly, since mobility ishigher than in MANETs, a fast medium access is neededin order to establish rapid communications among nodes.In this sense, the IEEE 802.11p MAC protocol is preferredrather than the traditional IEEE 802.11a/b/g. In IEEE 802.11p,several levels of prioritization are defined. Secondly, inVANETs scenarios, there are also static nodes of the network,namely, Road Side Units (RSUs), which are intended tooffer services to vehicles. Consequently, in VANET scenarios,Vehicle To Infrastructure communications (V2I) and VehicleTo Vehicle communications (V2V) can be distinguished.However, in disaster scenarios, these road units are likelyto be destroyed or malfunctioned so V2V communicationsshould be considered as the main application of VANETs.Another important difference is related to power consump-tion. While in MANETs nodes use the wireless transceiversof smartphones designed for low consumption, in VANETsthewireless transceivers are self-charged by the cars batteries.Consequently, energy is not a key routing parameter inVANETs. Both broadcasting schemes and routing protocolsfor VANETs have been active research fields in the last decade[65]. The readers can refer to [66] for more informationon projects and standardization actions in the vehicularnetworking.

Regarding the wireless technology to be used in VANETs,they have been recently standardized into two technologies.The first one was mainly led by the Unites States of America(USA) and its Department of Transportation (USDOT);it was called Wireless Access for Vehicular Environments(WAVE) [7]. It is a protocol suite composed by the IEEE802.11p and the IEEE 1609 protocols. The second one wasmainly led by the European Union (EU); it was calledCommunications Access for Land Mobiles (ISO CALM).Both protocols present similarities. The ISO CALM protocolallows more transmission technologies (GSM, UMTS, CENDSRC, etc.) having also includedWAVE as one of themunderthe name CALM-M5. The main shortcoming is that WAVEand ISO CALM technologies are still under study and theyare not used in commercial cars yet. In order to be used indisaster scenarios, public vehicles such as ambulances andpolices vehicles must be equipped withWAVE or ISO CALMtechnologies.

2.2.3. DTNs. Like MANETs, opportunistic networks are alsosuitable for disaster scenarios, for example, in scenario wherehigh density of nodes cannot be guaranteed or when theymove with high mobility. These conditions may cause amalfunction of the normal mechanism of an ad hoc routingprotocol (route discovery plus route establishment). InDTNs,we cannot make the distinction between broadcasting androuting protocol-based communications, because there isonly one way of communication between nodes, which isused in every new encounter between two nodes [67]. InDTN communication, the information is sent in units calledBundles. When a node generates information, it is split indifferent bundles and then the node waits until encounteringanother node in order to deliver the information (bundleprotocols). Consequently, while MANET routing protocolswork on network layer, the bundle protocols for DTNs work

on an upper layer, namely, bundle layer, which is betweenthe transport layer and application layer. Note that the DTNparadigm refers to the idea of storing and carrying theinformation until a new encounter occurs so we can alsoconsider DTNs in vehicular scenarios. In contrast toMANETscenarios, when a node cannot transmit a message to thedestination it puts the message in a sending buffer, and,after waiting for certain time, the node will drop the storedmessage.

As a rule,DTNparadigm is suitable in cases of lowdensityof nodes. For example, a police officer who is participating ina rescue operation and carrying certain information sharesthis information with all encountered victims. Furthermore,DTN are also suitable in high mobility networks, wheretraditional MANETs fail due to the continuous breakages ofroutes between the pair source-destination nodes.

Finally, the wireless technologies applicable for DTNs arethe same for MANETs, so WiFi and Bluetooth are the maincandidates.

2.2.4.WSNs. WSNs are suitable for early detection of possibledisasters, for example, earthquake detection systems andflooding detection. In [58], the authors state the main issuesto be solved when using WSN in disaster scenarios suchas discovery and naming, robust routing, prioritization ofcritical data, security, and tracking device location. Further-more, WSNs are envisioned to be an important componentto achieve the Internet of Things paradigm [5, 68, 69]. Con-sequently, interoperability with WSNs is a key requirementfor a disaster response communication system. WSNs arecentralized networks where nodes are normally grouped inclusters [53]. In general, a node acts as the central node, headof cluster, or sink node and the other nodes are collectingcertain data from the environment and sending it towards thecentral node. In this type of ad hoc network, it is possible toroute the data from a sensor node to the central node througha multihop communication path. However, there are cleardifferences from the previous multihop mechanisms. Firstly,power consumption is the key parameter when designingrouting protocols. Nodes in WSN are normally powered bybatteries, whose replacements or changes may be difficult oreven impossible. Secondly, wireless technology is focused onlow power consumption rather than on bandwidth.The IEEE802.15.4 standard has arisen as the standard de facto inWSNs,which allows nodes to communicate using a low rate butwith long networks lifetime. Another important differencefrom previous ad hoc networks is related to mobility ofnodes; in WSNs, mobility is very low since most nodes arestatic. As for the applicability in disaster scenarios, WSNsshould play detection and warning roles. WSN is a suitablecommunication paradigm to be deployed in areas that arelikely to suffer from natural disasters. Let us imagine a WSNdeployed in an area in order to sense the earth vibrations fordetecting possible earthquakes or aWSN deployed in a forestto monitor and detect fires.

2.2.5. WMNs. Regarding WMNs [54], the main commu-nications are similar to that of MANET paradigm (broad-casting and routing protocols). However, the applications of


traditional MANET and WMNs are different [54], makingthe goals of the routing protocols designed for WMNs alsodifferent. In WMNs, communications are mostly focused onextending Internet connectivity. We can imagine nodes inan urban area which have Internet connectivity th

Documents

Ad Hoc Networks for Disaster Scenarios and/or Threat Detectiondownloads.hindawi.com/journals/specialissues/967324.pdf · Ad Hoc Networks for Disaster Scenarios and/or Threat Detection